Data Storytelling Techniques: How to Tell a Great Data Story in 4 Steps

The image below shows a data storytelling example that uses bar charts and color design to highlight the main characters of a story about graduate rates for bachelor degrees. The graph on the left shows general information about graduate rates for bachelor degrees from the National Center for Education Statistics. When looking at the original data graphic, it takes a while to realize that the most meaningful information in the graph is that the private for-profit institutions have a much lower graduation rate. 

The optimized graph on the right shows the exact same data, but instead uses data storytelling techniques to create a clear narrative that makes the most important information stand out.  The main character of the data story is the high drop-out rate at private for-profit institutions compared to the other types of public and private institutions, so this has the boldest color that stands out from the rest of the data visualization. The improved graphic also simplifies the storyline by making the dataset only represent the change between graduated versus drop-out rates of all students, instead of using separate bars for different genders that had relatively similar graduation rates. By creating a clear story and using good color design, the dataset was transformed from a confusing bar graph into a clear data visualization where the main point is memorable and easy to understand.

Good stories contain four main elements: characters, setting, conflict, and resolution. In order to make a good data story, you can use these elements to frame the complex information into compelling visualizations. It may seem odd to think about your data points as characters in a story, but the steps below outline a proven technique to make your data story stand out from the crowd.

Create a short list of the 2-4 most important characters in your data story. These could be proteins, medications, or even methods. Think of the main data points or information as as heroes, villains, and supporting characters for the story.

For example, let's say that a biotechnology company creates a new product that improves an antibody purification process. The three main characters of this story would be the new biotech product, the competitor products, and the antibodies that become purified in the process. The hero is the biotech product, the villain is the competitor products, and the antibodies are the supporting characters.

The setting of a data story could be the methods, the scientific tools used, or the scientific background information. Use the setting as the scaffolding that allows your audience to make sense of the story but doesn’t play a major role in the outcome.

With the biotechnology example, the setting is the purification process that creates the hero and the villain characters. The biotech company product simplifies the process so that there is only one step to create the new natural nanoparticle product instead of three steps, and makes the process less toxic.

In data storytelling, the resolution is a synonym for the results. Everything in the story is leading to the main result that you want your audience to understand at the end.

The resolution for our example data story is the better performance of the biotech company's product. Taking these storytelling elements into account, we can now design a full infographic with illustrations and a graph that highlights this biotech product hero’s data story (see below).

• Source for National Center for Education Statistics data: https://nces.ed.gov/fastfacts/display.asp?id=40